Modelling, simulation, and analysis of supply chain systems using discrete-event simulation
Many approaches have been developed which support the construction of detailed supply chain models useful for analysis and simulation. However, most of these approaches lack the ability to model the supply chain in a single model, and usually produce solutions that lead to conflicting strategies between the companies. Simulation using a discrete-event simulation (DES) is an effective tool for the dynamically changing supply chain variables, thus allowing the system to be modelled more realistically. Considering the complexities of the supply chain system and the interrelations between its various systems, the task of developing such a model is challenging. The aim of this thesis is to develop a simulation model of a fast moving consumer goods (FMCG) supply chain with a DES tool. This model would be utilised as a decision-support system (DSS) for the investigation of the effectiveness of several inventory policies towards effective coordination and control of production inventory system, in various situations. This thesis discusses fundamental issues in the development of a simulation model for a supply chain using the DES tool, ARENA. A modelling procedure for the development of a supply chain simulation model is presented. The overall structure of the model is constructed by incorporating the well documented concept of modelling materials flowing downstream with an approach of modelling orders flowing upstream (modelling of feedback information). The model has an easily adaptable structure where rules (inventory policies) and model variables can be modified. The flexibility in the model's structure allows devising appropriate experimental designs, for several tests to be performed to imitate some realistic situations or scenarios (including the presence of disturbances). A new control theory oriented inventory policy, called the pseudo PID, is proposed. Detailed evaluations of five inventory policies for a production-inventory control under dynamic and stochastic conditions is presented. The findings demonstrate the ability of the approach to provide a wealth of potential solutions to the decision-maker, and confirm the qualitative behaviour of a supply chain in response to the different policies.